Parameter identification of permanent magnet synchronous machine based on metaheuristic optimization

Understanding the significance of precise dynamic modeling of electrical machines and the importance of parameter determination for the same, this manuscript proposes a new method of identifying variable inductances and damper parameters of a line-start interior permanent magnet synchronous machine (LSIPMSM) through an off-line improved particle swarm optimization (IPSO). An improved dynamic machine model incorporating the dependence of inductances on magnetizing currents has been developed. Through the combination of experimental test methods conducted on the inverter connected LSIPMSM under varied operating conditions and IPSO algorithm, parameters such as stator and magnetizing inductances and damper parameters have been identified for all conditions. Though conducted on LSIPMSM, the modeling and identification procedures presented in this paper are also applicable to IPMSM and surface magnet PSM with simplified variations. Comparison results of experiments with conventional and improved models are also presented for validation.